The transaqueductal pathway to the fourth ventricleshould not be considered analogous to the Pillars of Her-cules, beyond which one dares not navigate.10 On review-ing the historical experience, we noted that the very fewpioneers of endoscopic approaches to the aqueductal regionby a frontal route preferred rigid instruments, and that theextraordinary versatility of flexible scopes has not beenfully exploited to achieve complete control of the ventricu-lar cavities.3,6,7,9

In this study we summarize our findings in a series of 54patients presenting with different pathological conditions inwhich transaqueductal navigation has been successfullyperformed using flexible endoscopes. This instrument hasallowed us to apply this approach, not as a challenge to thestatus quo, but as a surprisingly similar method to thosecommonly used in the more usual neuroendoscopic proce-dures.

TECHNICAL PROCEDURE

The surgical approach is performed through a frontal burhole 2 cm anterior to the coronal suture and 2 cm lateral tothe midline. The direction of the trajectory needed to reachthe ventricles is the same as that used to perform a third ven-triculostomy. The frontal horn is first cannulated with a No.14 French peel-away catheter, and through that access porta flexible endoscope is introduced. During the whole proce-dure, the instrument is managed with a freehand techniqueby using movements of advancement, flexion, extension,and rotation. In the third ventricle, the scope is directedtoward the opening of the cerebral aqueduct and advancedinto the ampulla and then toward the fourth ventricle.

Although it is a confined space, the cerebral aqueduct isendowed with its own compliance;4 the aqueductal channelyields at the passage of an instrument 4 mm in width. Afterthe endoscope crosses the opening of the cerebral aqueduct(Fig. 1), an initial narrowing, corresponding to the superiorcolliculi, is recognizable, and then a more dilated spacecalled the aqueductal ampulla is encountered.1,4,10 A furtherrestriction is found corresponding to the inferior colliculi.The aqueductal space then opens under the frenulum of theposterior medullary velum into the rhomboid fossa.

In the navigation of the fourth ventricle some precau-tions should be adopted. Once the aqueduct is cannulated,irrigation is stopped because the instrument itself occupiesand shuts the passage with its own volume. Further addi-tions of liquid could overload the fourth ventricle, which inthis particular phase of the operation could become com-pletely trapped (depending on the underlying pathologicalentity). Related episodes of threatening bradycardia couldbe the clinical consequence.5

Another observation concerns the view offered to theneurosurgeon; the posterior deflection of the scope and pos-terior rotation cause an upside-down view of the video en-doscopic images. Therefore, it should be kept in mind thatthe dorsal and ventral anatomical structures are projecteddownward and upward, respectively, in the monitor. In thatway, for instance, when the floor of the rhomboid fossa isup, the fastigium is down. Also, the tela choroidea with theaffixed plexus is visible in the downward position of thevideoendoscopic monitor (Fig. 2).

The sulcus medianus of the fourth ventricle is the guid-ing anatomical mark leading toward the posterior triangleof the rhomboid fossa, with clear evidence of calamus

Neurosurg Focus 19 (6):E12, 2005

Endoscopic navigation of the fourth ventricle

Technical note and preliminary experience

PIERLUIGI LONGATTI, M.D., LUCA BASALDELLA, M.D., ALBERTO FELETTI, M.D.,ALESSANDRO FIORINDI, M.D., AND DOMENICO BILLECI, M.D.

Department of Neurosurgery, Treviso Regional Hospital, University of Padova, Italy

Transaqueductal navigation of the fourth ventricle has long been considered dangerous and of no clinical relevance.After the refinement of the endoscopic technique and supported by the extensive experience gained at the authors’ in-stitution since 1994, endoscopic exploration of the fourth ventricle has been performed by the same surgeon in 54 pa-tients. In all cases reviewed, endoscopic navigation of the fourth ventricle was successfully performed with no relatedneurological deficit. This preliminary experience shows the feasibility of transaqueductal navigation of the fourth ven-tricle, which is made possible by the use of small, flexible endoscopes in expert hands.

KEY WORDS • neuroendoscopy • fourth ventricle • cerebral aqueduct • ventricleanatomy • hydrocephalus

Neurosurg. Focus / Volume 19 / December, 2005 1

Unauthenticated | Downloaded 06/16/21 07:15 AM UTC

scriptorius and the foramen of Magendie (Fig. 3). In thefourth ventricle, the sulcus medianus acts as a route to thecalamus scriptorius, but generally the instrument’s ad-vancement causes a physiological deflection toward thecontralateral left lateral recess, then to the left foramen ofLuschka. Navigation of the posterior triangle of the rhom-boid fossa is possible, although it is anatomically delimitedby the striae medullares ventriculi quarti. Nevertheless, it issufficient to rotate the endoscope until crossing the midline,then reaching toward the calamus scriptorius and the obexabove, which contain the canalis centralis medullaris spi-nalis.

Cerebellar tonsils and posterior inferior cerebellar arter-ies surround and delineate the dorsal part of the foramen ofMagendie, which, when patent and wide, allows a view tothe posterior oblongata and posterior part of the foramenmagnum outside the fourth ventricle. Inspection of the for-amen of Luschka is obtained by simply rotating the endo-scope a little with moderate flexion. When withdrawing thescope, it is always advisable to look into the opening of thecerebral aqueduct to notice possible signs of small ependy-mal contusions.

PATIENT SERIES

More than 300 endoscopic procedures have been per-formed between 1994 and 2005 at our institution, and theseprocedures include 54 endoscopic navigations of the fourthventricle. We reviewed our cases by retrieving data fromboth our neuroendoscopic clinical and video databases. Theclinical series was subdivided into six clinicopathologicalgroups, as follows. Group 1, five patients with a trapped

fourth ventricle who underwent aqueductoplasty and stentinsertion, followed by exploration of fourth ventricle out-lets. Group 2, 19 patients harboring apparently communi-cating normal-pressure hydrocephalus treated with endo-scopic third ventriculostomy, in whom navigation of thefourth ventricle was performed to rule out possible closingof outlets. Group 3, five patients with triventricular hydro-cephalus due to membranous occlusion of the fourth ventri-cle. Group 4, 20 patients with intraventricular hemorrhageswho underwent endoscopic ventricular clot aspiration,including the fourth ventricle. Group 5, three cases of aque-ductal exophytic soft gliomas floating toward the fourthventricle, which were directly aspirated. Group 6, two cases(one colloid cyst and one cystic intraventricular craniophar-yngioma) in which the fourth ventricle was explored to as-pirate fragments of colloid or cholesteremic cyst contents.

Our series is heterogeneous, but we tried to identify thedifficulties and the feasibilities of the technical procedurefor each case reviewed. Neuroendoscopic navigation of thefourth ventricle was successfully performed in all cases; werecorded four episodes of extreme bradycardia, and smallependymal contusions of the opening of the cerebral aque-duct in two cases. Possible brain damage due to navigation-related injury is difficult to see in the neuroendoscopic as-sessment of intraventricular hemorrhages because of thepresence of clots; besides, in chronic hydrocephalus theaqueduct could be so dilated that no resistance is offered tothe introduction of an endoscope.

DISCUSSION

Authors of few reports have focused on the specific issue

P. Longatti, et al.

2 Neurosurg. Focus / Volume 19 / December, 2005

Fig. 1. A computer-modified photograph of a plastic model showing the location of an axial section of the brainstemat the opening of the cerebral aqueduct (left) and a neuroendoscopic view of the anatomy (right).

Unauthenticated | Downloaded 06/16/21 07:15 AM UTC

http://www.aans.org/education/journal/neurosurgical/dec05/19-6-12f1.html

of endoscopic exploration of the fourth ventricle. Matula, etal.,7 stated that there are three possible ways to reach thefourth ventricle endoscopically: 1) coming from the thirdventricle via the cerebral aqueduct; 2) through the basal cis-tern opening of the lateral foramen of Luschka; and 3) viathe cerebellomedullary cistern through the foramen of Ma-gendie.

The history of neuroendoscopic exploration of the fourthventricle has been written by neurosurgeons using rigidendoscopes rather than by those adopting the more appro-priate flexible ones. Endoscopic approaches to the aque-ductal region via a frontal route were performed by Schroe-der and Gaab,9 who used rigid endoscopes for the treatmentof aqueductal stenosis in 17 patients. These authors ob-served postoperative sequelae mainly associated with dis-turbances of ocular movement.

Anandh, et al.,1 described a right transfrontal approach inwhich a rigid endoscope was used to enter the lateral andthird ventricle to remove fourth ventricle cysts in three pa-tients. Cystic lesions were removed in all three with nodeaths, although one patient experienced transient hemi-paresis and ocular ptosis. In a patient with fourth ventricleneurocysticercosis, Proano, et al.,8 positioned a rigid endo-scope in front of the opening of the cerebral aqueduct andaspirated the cysts through a No. 6 French catheter.

Experience with the flexible endoscope for exploration ofthe fourth ventricle in five cases was reported by Bergsnei-der.3 He avoided the transaqueductal pathway, however, andpreferred a midline durotomy between the opisthion and theposterior arch of C-1, advancing toward the fourth ventriclethrough the foramen of Magendie.

Neuronavigation beyond the aqueduct can be of some in-terest in carefully selected situations.2 Our preliminary ex-

periences commenced with cases of trapped fourth ventri-cles in infantile posthemorrhagic hydrocephalus. In thesecases, the surgical action consisted of the shearing of theaqueductal membrane, which was further dilated with cau-tious partial inflation of a particular Fogarty balloon. Ex-ploration proved to be very difficult because of the enlargedventricle and poor visualization of the rhomboid fossa; inmost of the cases this structure was so deformed that it hadbecome unrecognizable.

Another interesting indication for fourth ventricle explo-ration is the total removal of intraventricular hemorrhagesdetailed in previous reports.6 What we learned is the ease ofclearing clots from the fourth ventricle with this procedurecompared with others. Also, the time spent varied greatly: afew minutes in the fourth ventricle and approximately 1hour to clear each lateral ventricle. This unexpected ease indealing with the fourth ventricle navigation was further con-firmed by postoperative follow-up neuroimages, which re-vealed the fourth ventricle to be completely clot free. Onceagain, the anatomical integrity, when exiting the opening ofthe cerebral aqueduct, revealed no signs of damages relatedto the endoscopic maneuvers.

Many theoretical objections could be raised against thetransaqueductal navigation of the fourth ventricle, but inlight of our experience, this procedure was found to be rel-atively easy and harmless. The only neurological compli-cation observed in our series was a mild diplopia in apatient in whom aqueductoplasty was followed by patencyof the fourth ventricle outlets. The neurological deficit wasalmost certainly caused by the aqueductoplasty rather thanby the transaqueductal passage of the endoscope.

The previously reported experiences of extreme brady-cardia due to excessive irrigation5 have taught us to be

Neurosurg. Focus / Volume 19 / December, 2005

Endoscopic navigation of the fourth ventricle: technical note

3

Fig. 2. Specimen showing the location of an axial section of the brainstem at the superior triangle of the rhomboidfossa in the fourth ventricle (left) and neuroendoscopic visualization of the ventricular anatomical structures (right).

Unauthenticated | Downloaded 06/16/21 07:15 AM UTC

http://www.aans.org/education/journal/neurosurgical/dec05/19-6-12f2.html

much more cautious and accurate in dealing with the fourthventricle and possible hypertension. Even if episodes of asignificant bradycardia occurred, they immediately regres-sed after the withdrawal of small volumes of ventricularfluid.

CONCLUSIONS

We think that transaqueductal exploration of the fourthventricle is feasible and reasonably safe, provided that thesurgeon has extensive experience in neuroendoscopy per-formed using flexible instruments managed with the free-hand technique.

References

1. Anandh B, Mohanty A, Sampath S, et al: Endoscopic approachto intraventricular cysticercal lesions. Minim Invasive Neuro-surg 44:194–196, 2001

2. Barkhof F, Kouwenhoven M, Scheltens P, et al: Phase-contrastcine MR imaging of normal aqueductal CSF flow. Acta Radiol35:123–130, 1994

3. Bergsneider M: Endoscopic removal of cysticercal cysts withinthe fourth ventricle. Technical note. J Neurosurg 91:340–345,1999

4. Flyger G, Hjelmquist U: Normal variations in the caliber of thehuman cerebral aqueduct. Anat Rec 127:151–162, 1957

5. Handler MH, Abbott R, Lee M: A near-fatal complication ofendoscopic third ventriculostomy: case report. Neurosurgery35:525–528, 1994

6. Longatti PL, Martinuzzi A, Fiorindi A, et al: Neuroendoscopicmanagement of intraventricular hemorrhage. Stroke 35:35–38,2004

7. Matula C, Reinprecht A, Roessler K, et al: Endoscopic explorationof the IVth ventricle. Minim Invasive Neurosurg 39:86–92,1996

8. Proano JV, Madrazo I, Avelar F, et al: Medical treatment forneurocysticercosis characterized by giant subarachnoid cysts. NEngl J Med 345:879–885, 2001

9. Schroeder HW, Gaab MR: Endoscopic aqueductoplasty: tech-nique and results. Neurosurgery 45:508–518, 1999

10. Woollam DHM, Millen JW: Anatomical considerations in the aethiology of stenosis of the cerebral aqueduct. Brain 76:104–112, 1953

Manuscript received October 20, 2005.Accepted in final form November 21, 2005.Address reprint requests to: Luca Basaldella, M.D., Divisione

Clinicizzata di Neurochirurgia, Ospedale Regionale di Treviso, Ca’Foncello, Facoltà di Medicina e Chirurgia, Università degli Studi diPadova, Piazza le Ospedale Civile 1, 31100 Treviso, Italia. email:lbasaldella@ulss.tv.it.

P. Longatti, et al.

4 Neurosurg. Focus / Volume 19 / December, 2005

Fig. 3. Specimen showing the location of an axial section of the brainstem at the inferior triangle of the rhomboid fos-sa in the fourth ventricle (left) and neuroendoscopic visualization of the ventricular anatomical structures (right).Magendie = foramen of Magendie.

Unauthenticated | Downloaded 06/16/21 07:15 AM UTC

http://www.aans.org/education/journal/neurosurgical/dec05/19-6-12f3.html